Background We have shown that NRL interacts with TATA-binding protein (TBP), CRX, NR2E3, SP4 and other transcriptional regulatory factors to control gene expression. Gene profiling of Nrl-KO mouse retina demonstrated that NRL affects the expression of a large group of genes in rod photoreceptors, including rhodopsin. Furthermore, deletion of Nrl in mouse leads to functional transformation of rods to cone photoreceptors, whereas expression of Nrl in committed cone precursors can convert these cells to rods. Understanding the molecular networks that regulate NRL and/or are regulated by NRL and functional characterization of individual key components could provide fundamental insights into photoreceptor cell biology and dysfunction. Results 1. Spatio-temporal gene profiling of rod photoreceptors By adopting a high throughput approach, we are using photoreceptor gene profiling to construct gene expression networks underlying functional maturation and homeostasis of rods. We take advantage of a line of transgenic mice, which we generated and which expresses green fluorescent protein (GFP) in newly born and mature rod photoreceptors under control of the NRL-promoter. Over sixty RNA samples have been successfully generated from Nrl-GFP positive, flow-sorted mouse photoreceptor cells, representing several time points for gene profiling of developing photoreceptors, of mature photoreceptors under light-dark cycles, in aging and disease conditions. The latter two projects will also allow us to examine how gene expression changes in aging and diseased photoreceptors affect cellular function and visual transduction (see project # EY00475-01). 2. NRL downstream pathways We are employing in silico and chromatin immunoprecipitation (ChIP) followed by high throughput sequencing (ChIP-seq) methods to identify direct targets of NRL and of other retinal transcription factors. NRL ChIP-seq analyses, combined with Nrl-KO gene profiles, have identified many direct transcriptional targets of NRL in mature mouse rod photoreceptors. ChIP-seq analysis was also validated with ChIP-qPCR analysis. Functional roles of NRL target genes are being tested by conducting in vivo RNAi-mediated knockdown studies in the developing mouse retina. Conditions for NRL P2 ChIP and CRX adult ChIP have been established to allow further study the targets of these two important transcription factors in developing and mature photoreceptors. Furthermore, to understand the molecular basis of rod-specific gene regulation we have isolated from the bovine retina a high molecular mass protein complex containing NRL. Preliminary analysis has shown co-elution of known interactors (e.g. NR2E3, CRX, and Pol II) with NRL. More in depth analysis of NRL-interacting proteins contained in the high molecular mass complex is being conducted by immunoprecipitation (IP) using NRL IgG followed by WB analysis with candidate antibodies and mass spectrometry. We are evaluating, by in vivo electroporation and in vitro, Nrl candidate targets identified by ChIP-seq and microarrays. A group of promising targets will be extensively characterized. NRL is the key transcriptional regulator of the rod photoreceptor-specific gene rhodopsin. To characterize the rhodopsin enhanceosome, different streptavidin tagged oligonucleotides have been designed from the rhodopsin promoter and enhancer region. Oligonucleotide affinity chromatography is being performed from the bovine retinal nuclear fraction to isolate individual components binding to different regions of the rhodopsin promoter. Analysis of the different components bound to oligonucleotides is in progress using candidate antibodies and mass spectrometry. 3. NRL upstream pathways To understand the regulation of Nrl expression during photoreceptor differentiation and in the mature retina, we are investigating cis-elements and regulatory binding factors in Nrl promoter. We are also examining post-translational modifications that modulate NRL activity. In vivo electroporation in the mouse retina was adopted to test Nrl enhancer/promoter-reporter constructs. Nrl promoter sequences spanning one to four transcriptional binding site clusters were compared for their capacity to drive expression of GFP. Initial results showed that 2 clusters corresponding to 900 bp upstream sequence of the Nrl promoter are sufficient to sustain reporter gene expression specifically in rod photoreceptors. These binding sites would suggest a role for bHLH proteins and nuclear receptors in regulating Nrl transcription. To delineate post-translational modifications that control the activity of NRL, we have investigated SUMOylation and phosphorylation of NRL and the corresponding signaling pathways. We have shown that di-SUMOylation of lysine at position 20 of NRL protein occurs in adult retinas in vivo and that it modulates NRL transcriptional activity on rhodopsin and Nr2e3 promoters in vitro. Sumoylation of NRL in developing photoreceptors is important to complete their differentiation programs as shown by the inability of mutant Nrl constructs lacking SUMOylation sites to rescue rod photoreceptor cell differentiation defects in Nrl-KO mice after in vivo electroporation in neonatal retina. Furthermore, preliminary data suggest that PIAS3, a transcriptional inhibitor that interacts with NR2E3 and CRX and is essential for photoreceptor differentiation, interacts also with NRL and might be involved in the SUMOylation of NRL, CRX and NR2E3. To investigate the role of this protein in modulating NRL function and retinal development, we have generated PIAS3 floxed mice. These mice will be bred with Rx-Cre (for deletion in all retinal cells) and with Crx-Cre (for photoreceptor specific deletion) to establish the role of SUMOylation in regulating NRL function in specifying photoreceptor cell fate. GSK3-&#946;and MAPKs have been suggested to play important roles in NRL phosphorylation and in controlling its activity. Our preliminary results showed that GSK3&#946;is expressed in cones. We are currently investigating GSK3&#945;expression. Conditional deletion of GSK3&#945;in all retinal progenitors led to several retinal aberrations. Double GSK3&#945;-GSK3&#946;KO mice are currently being crossed for further investigation. 4. Effects of NRL mutations on photoreceptor development and homeostasis The transgenic mice that we generated to express human NRL S50T and NRL P51S mutations in Nrl-KO mice underwent normal retinal development and rod differentiation. No abnormal histological or functional (ERG) phenotype could be observed up to 12 months of age. However, upon short exposure (1 hr) to bright light almost complete rod degeneration was observed even at young ages. No damage was observed in the cone population. Similar experiments are in progress with NRL S50T transgenic mice. S50T and P51S mutations prevent NRL phosphorylation and increase its transcriptional activity on the rhodopsin promoter. However, it would appear that NRL phosphorylation is dispensable for rod differentiation but is essential for rod maintenance. Significance As our studies identify novel genes and networks that characterize photoreceptor development and function we gain new insights into photoreceptor biology and disease and can make more accurate predictions on biomarkers and candidate therapeutic targets as well as preventive measures (e.g., patients carrying NRL S50T and/or P51S mutations might benefit from wearing dark sunglasses to protect their rods from bright light exposure and consequent damage)